# !pip install --trusted-host pypi.org --trusted-host pypi.python.org --trusted-host files.pythonhosted.org -U kaleido

Metrolinx Rideship Regression Analysis

# Importing all necessary libraries to proceed with this project.
import warnings
warnings.filterwarnings("ignore")

import pandas as pd
import numpy as np
import scipy.stats as stats
import seaborn as sns
import matplotlib.pyplot as plt
import calendar
import random
import sqlite3
import itertools
from itertools import cycle, islice
from datetime import timedelta

import plotly.express as px
from sklearn.model_selection import cross_val_score, cross_val_predict
from sklearn import metrics
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import Lasso, LinearRegression, LassoCV, Ridge
from sklearn.model_selection import train_test_split, cross_val_score, cross_validate, GridSearchCV 
from sklearn.neighbors import KNeighborsRegressor
from sklearn.experimental import enable_hist_gradient_boosting
from sklearn.pipeline import make_pipeline

import os
ss = StandardScaler()

import utils
from utils import adjustedR2

os.getcwd()

'C:\\Users\\DKici\\Desktop\\Metrolinx\\notebook'

Load and read data

input_path = f"{os.getcwd()}\\input_data\\"
ts_path = f"{os.getcwd()}\\ts_data\\"
output_path = "C:\\Users\\DKici\\Desktop\\Metrolinx\\"
figures_path = f"{output_path}\\figures\\"

figures_path

'C:\\Users\\DKici\\Desktop\\Metrolinx\\\\figures\\'

rc_df = pd.read_csv(f"{ts_path}\\Rail_Corridor_Data.csv")
rc_df = rc_df.fillna(0)
plt.figure(figsize = (8,8))
ax = sns.heatmap(rc_df.corr(),annot=True, cbar=False, cmap='Blues', fmt='.1f')

st_df = pd.read_csv(f"{ts_path}\\Station_Data.csv")
st_df = st_df.fillna(0)

stations = ['Barrie','Kitchener','Lakeshore East','Lakeshore West', 'Milton', 'Richmond Hill', 'Stouffville']

for st in stations:
    print(st)
    st_temp =  st_df.columns[st_df.columns.str.contains(pat = f'{st}')] 

    plt.figure(figsize = (15,15))
    ax = sns.heatmap(st_df[st_temp].corr(),annot=True, cbar=False, cmap='Blues', fmt='.1f')
    plt.savefig(f'{figures_path}/{st}_stations_correlations.png')

Barrie
Kitchener
Lakeshore East
Lakeshore West
Milton
Richmond Hill
Stouffville

rc_df["date"]=pd.to_datetime(rc_df.date)
rc_df["Week"]=rc_df.date.dt.week
rc_df["Month"]=rc_df.date.dt.month
rc_df["Year"]=rc_df.date.dt.year
# Changing the Months value from numbers to real values like Jan, Feb to Dec
# rc_df['Month'] = rc_df['Month'].apply(lambda x: calendar.month_abbr[x])

rc_df.describe().T

	count	mean	std	min	25%	50%	75%	max
ridership Barrie	1461.0	4400.605065	5347.616399	0.0	757.0	1813.0	5768.0	19119.0
ridership Kitchener	1461.0	4856.924709	6199.820190	0.0	0.0	1647.0	7615.0	23991.0
ridership Lakeshore East	1461.0	12375.724162	11884.308002	0.0	3421.0	7422.0	17304.0	50318.0
ridership Lakeshore West	1461.0	16216.637919	15623.745199	0.0	3679.0	9839.0	25178.0	71920.0
ridership Milton	1461.0	4897.070500	7363.907604	0.0	0.0	946.0	5430.0	24170.0
ridership Richmond Hill	1461.0	1880.291581	3009.700675	0.0	0.0	244.0	1807.0	8978.0
ridership Stouffville	1461.0	3457.464750	4609.707402	0.0	295.0	928.0	4520.0	17103.0
total_train_trips Barrie	1461.0	25.484600	10.512123	0.0	21.0	29.0	34.0	56.0
total_train_trips Kitchener	1461.0	23.981520	17.237235	0.0	0.0	30.0	33.0	86.0
total_train_trips Lakeshore East	1461.0	80.175222	31.562974	0.0	66.0	73.0	118.0	232.0
total_train_trips Lakeshore West	1461.0	78.433265	27.903733	0.0	68.0	79.0	106.0	218.0
total_train_trips Milton	1461.0	9.458590	7.610859	0.0	0.0	10.0	19.0	40.0
total_train_trips Richmond Hill	1461.0	6.167009	4.613207	0.0	0.0	7.0	11.0	24.0
total_train_trips Stouffville	1461.0	23.829569	11.906098	0.0	22.0	28.0	32.0	60.0
Week	1461.0	26.626968	15.069603	1.0	14.0	27.0	40.0	53.0
Month	1461.0	6.522930	3.449884	1.0	4.0	7.0	10.0	12.0
Year	1461.0	2020.499658	1.118111	2019.0	2020.0	2020.0	2021.0	2022.0

len(st_df.columns) # 59 stations in total

117

st_df["date"]=pd.to_datetime(st_df.date)
st_df["Week"]=st_df.date.dt.week
st_df["Month"]=st_df.date.dt.month
st_df["Year"]=st_df.date.dt.year
# Changing the Months value from numbers to real values like Jan, Feb to Dec
st_df['Month'] = st_df['Month'].apply(lambda x: calendar.month_abbr[x])
st_df.describe().T

	count	mean	std	min	25%	50%	75%	max
ridership Barrie ALLANDALE WATERFRONT GO	1461.0	1743.847365	2200.572140	0.0	265.0	567.0	2509.0	7502.0
ridership Barrie AURORA GO	1461.0	260.076660	217.992864	0.0	57.0	218.0	438.0	1393.0
ridership Barrie BARRIE SOUTH GO	1461.0	0.560575	21.426873	0.0	0.0	0.0	0.0	819.0
ridership Barrie BRADFORD GO	1461.0	237.510609	391.559584	0.0	0.0	55.0	134.0	2050.0
ridership Barrie KING CITY GO	1461.0	0.045175	1.726708	0.0	0.0	0.0	0.0	66.0
...	...	...	...	...	...	...	...	...
total_train_trips Stouffville STOUFFVILLE GO	1461.0	0.000684	0.026162	0.0	0.0	0.0	0.0	1.0
total_train_trips Stouffville UNION STATION	1461.0	11.960301	6.001124	0.0	11.0	14.0	16.0	30.0
total_train_trips Stouffville UNIONVILLE GO	1461.0	0.262149	1.139153	0.0	0.0	0.0	0.0	12.0
Week	1461.0	26.626968	15.069603	1.0	14.0	27.0	40.0	53.0
Year	1461.0	2020.499658	1.118111	2019.0	2020.0	2020.0	2021.0	2022.0

118 rows × 8 columns

for st in stations:
    data = rc_df[[f'ridership {st}','Week','Month','Year']]
    data_weeks = data.groupby('Week').sum()

    fig = px.line(data_frame = data_weeks, x = data_weeks.index, y = f'ridership {st}', labels = {'ridership' : 'Weekly Sales', 'x' : 'Weeks' }, title = f'ridership {st} over weeks')
    fig.update_traces(line_color='deeppink', line_width=3)
    fig.show()
    fig.write_image(f'{figures_path}/ridership {st}.png')

for st in stations:
    st_temp =  st_df.columns[st_df.columns.str.contains(pat = f'{st}')]
    for st_num in st_temp:
        print(st_num)
        data = st_df[[f'{st_num}','Week','Month','Year']]
        data_weeks = data.groupby('Week').sum()

        fig = px.line(data_frame = data_weeks, x = data_weeks.index, y = f'{st_num}', labels = {'ridership' : 'Weekly Sales', 'x' : 'Weeks' }, title = f'ridership {st} over weeks')
        fig.update_traces(line_color='deeppink', line_width=3)
        fig.show()
        fig.write_image(f'{figures_path}/{st_num}.png')

ridership Barrie ALLANDALE WATERFRONT GO

ridership Barrie AURORA GO

ridership Barrie BARRIE SOUTH GO

ridership Barrie BRADFORD GO

ridership Barrie KING CITY GO

ridership Barrie MAPLE GO

ridership Barrie RUTHERFORD GO

ridership Barrie UNION STATION

total_train_trips Barrie ALLANDALE WATERFRONT GO

total_train_trips Barrie AURORA GO

total_train_trips Barrie BARRIE SOUTH GO

total_train_trips Barrie BRADFORD GO

total_train_trips Barrie KING CITY GO

total_train_trips Barrie MAPLE GO

total_train_trips Barrie RUTHERFORD GO

total_train_trips Barrie UNION STATION

ridership Kitchener BRAMALEA GO

ridership Kitchener GEORGETOWN GO

ridership Kitchener GUELPH CENTRAL GO

ridership Kitchener KITCHENER GO

ridership Kitchener LONDON VIA

ridership Kitchener MALTON GO

ridership Kitchener MOUNT PLEASANT GO

ridership Kitchener UNION STATION

total_train_trips Kitchener BRAMALEA GO

total_train_trips Kitchener GEORGETOWN GO

total_train_trips Kitchener GUELPH CENTRAL GO

total_train_trips Kitchener KITCHENER GO

total_train_trips Kitchener LONDON VIA

total_train_trips Kitchener MALTON GO

total_train_trips Kitchener MOUNT PLEASANT GO

total_train_trips Kitchener UNION STATION

ridership Lakeshore East AJAX GO

ridership Lakeshore East ALDERSHOT GO

ridership Lakeshore East BURLINGTON GO

ridership Lakeshore East DURHAM COLLEGE OSHAWA GO

ridership Lakeshore East GUILDWOOD GO

ridership Lakeshore East NIAGARA FALLS GO (VIA STATION)

ridership Lakeshore East OAKVILLE GO

ridership Lakeshore East OSHAWA GO

ridership Lakeshore East PICKERING GO

ridership Lakeshore East PICKERING SOUTH

ridership Lakeshore East UNION STATION

ridership Lakeshore East WEST HARBOUR

ridership Lakeshore East WHITBY GO

total_train_trips Lakeshore East AJAX GO

total_train_trips Lakeshore East ALDERSHOT GO

total_train_trips Lakeshore East BURLINGTON GO

total_train_trips Lakeshore East DURHAM COLLEGE OSHAWA GO

total_train_trips Lakeshore East GUILDWOOD GO

total_train_trips Lakeshore East NIAGARA FALLS GO (VIA STATION)

total_train_trips Lakeshore East OAKVILLE GO

total_train_trips Lakeshore East OSHAWA GO

total_train_trips Lakeshore East PICKERING GO

total_train_trips Lakeshore East PICKERING SOUTH

total_train_trips Lakeshore East UNION STATION

total_train_trips Lakeshore East WEST HARBOUR

total_train_trips Lakeshore East WHITBY GO

ridership Lakeshore West ALDERSHOT GO

ridership Lakeshore West APPLEBY GO

ridership Lakeshore West BRONTE GO

ridership Lakeshore West BURLINGTON GO

ridership Lakeshore West CLARKSON GO

ridership Lakeshore West EXHIBITION GO

ridership Lakeshore West HAMILTON GO CENTRE

ridership Lakeshore West MIMICO GO

ridership Lakeshore West NIAGARA FALLS GO (VIA STATION)

ridership Lakeshore West OAKVILLE GO

ridership Lakeshore West OSHAWA GO

ridership Lakeshore West PICKERING SOUTH

ridership Lakeshore West PORT CREDIT GO

ridership Lakeshore West ST. CATHARINES GO (VIA STATION)

ridership Lakeshore West UNION STATION

ridership Lakeshore West WEST HARBOUR

ridership Lakeshore West WEST HARBOUR GO

total_train_trips Lakeshore West ALDERSHOT GO

total_train_trips Lakeshore West APPLEBY GO

total_train_trips Lakeshore West BRONTE GO

total_train_trips Lakeshore West BURLINGTON GO

total_train_trips Lakeshore West CLARKSON GO

total_train_trips Lakeshore West EXHIBITION GO

total_train_trips Lakeshore West HAMILTON GO CENTRE

total_train_trips Lakeshore West MIMICO GO

total_train_trips Lakeshore West NIAGARA FALLS GO (VIA STATION)

total_train_trips Lakeshore West OAKVILLE GO

total_train_trips Lakeshore West OSHAWA GO

total_train_trips Lakeshore West PICKERING SOUTH

total_train_trips Lakeshore West PORT CREDIT GO

total_train_trips Lakeshore West ST. CATHARINES GO (VIA STATION)

total_train_trips Lakeshore West UNION STATION

total_train_trips Lakeshore West WEST HARBOUR

total_train_trips Lakeshore West WEST HARBOUR GO

ridership Milton LISGAR GO

ridership Milton MILTON GO

ridership Milton UNION STATION

total_train_trips Milton LISGAR GO

total_train_trips Milton MILTON GO

total_train_trips Milton UNION STATION

ridership Richmond Hill BLOOMINGTON GO

ridership Richmond Hill GORMLEY GO

ridership Richmond Hill UNION STATION

total_train_trips Richmond Hill BLOOMINGTON GO

total_train_trips Richmond Hill GORMLEY GO

total_train_trips Richmond Hill UNION STATION

ridership Stouffville MILLIKEN GO

ridership Stouffville MOUNT JOY GO

ridership Stouffville OLD ELM GO

ridership Stouffville STOUFFVILLE GO

ridership Stouffville UNION STATION

ridership Stouffville UNIONVILLE GO

total_train_trips Stouffville MILLIKEN GO

total_train_trips Stouffville MOUNT JOY GO

total_train_trips Stouffville OLD ELM GO

total_train_trips Stouffville STOUFFVILLE GO

total_train_trips Stouffville UNION STATION

total_train_trips Stouffville UNIONVILLE GO

for st in stations:
    data = rc_df[f'ridership {st}']
    plt.figure(figsize=(10, 6))
    sns.distplot(data, kde=False, bins=30, color = 'tomato')
    plt.title(f'Ridership {st} Distribution')
    plt.show()

for st in stations:
    st_temp =  st_df.columns[st_df.columns.str.contains(pat = f'{st}')]
    for st_num in st_temp:
        data = st_df[f'{st_num}']

        plt.figure(figsize=(10, 6))
        sns.distplot(data, kde=False, bins=30, color = 'tomato')
        plt.title(f'Ridership {st_num} Distribution')
        plt.show()

Regression Analysis

df = rc_df.iloc[:,1:] # reomove date column
df.head()

	ridership Barrie	ridership Kitchener	ridership Lakeshore East	ridership Lakeshore West	ridership Milton	ridership Richmond Hill	ridership Stouffville	total_train_trips Barrie	total_train_trips Kitchener	total_train_trips Lakeshore East	total_train_trips Lakeshore West	total_train_trips Milton	total_train_trips Richmond Hill	total_train_trips Stouffville	Week	Month	Year
0	1149.0	0.0	5408.0	7056.0	0.0	0.0	0.0	25.0	0.0	72.0	72.0	0.0	0.0	0.0	1	1	2019
1	10297.0	12586.0	25057.0	29978.0	14600.0	5940.0	9087.0	36.0	31.0	118.0	107.0	20.0	12.0	29.0	1	1	2019
2	11395.0	13476.0	27819.0	34463.0	15327.0	6139.0	9671.0	36.0	31.0	119.0	108.0	20.0	12.0	30.0	1	1	2019
3	9762.0	11701.0	25222.0	30343.0	13328.0	5396.0	8636.0	36.0	31.0	112.0	102.0	20.0	12.0	30.0	1	1	2019
4	1549.0	0.0	7751.0	11118.0	0.0	0.0	0.0	25.0	0.0	69.0	69.0	0.0	0.0	0.0	1	1	2019

# train-test 75-25 split
df_train, df_test = train_test_split(df, 
                                     train_size = 0.75, 
                                     test_size = 0.25, 
                                     random_state = 100)

# rescale the features
scaler = MinMaxScaler()

# apply scaler() to all the numeric columns 
numeric_vars = df.columns
df_train[numeric_vars] = scaler.fit_transform(df_train[numeric_vars])
df_train[numeric_vars].head()

# apply rescaling to the test set also
df_test[numeric_vars] = scaler.fit_transform(df_test[numeric_vars])
df_test[numeric_vars].head()

	ridership Barrie	ridership Kitchener	ridership Lakeshore East	ridership Lakeshore West	ridership Milton	ridership Richmond Hill	ridership Stouffville	total_train_trips Barrie	total_train_trips Kitchener	total_train_trips Lakeshore East	total_train_trips Lakeshore West	total_train_trips Milton	total_train_trips Richmond Hill	total_train_trips Stouffville	Week	Month	Year
1437	0.448967	0.556708	0.505077	0.512551	0.336072	0.263194	0.436526	0.888889	0.867925	0.635659	0.666667	0.6	0.583333	0.916667	0.923077	1.000000	1.000000
57	0.858954	0.866591	0.815689	0.710300	0.798310	0.861950	0.885164	1.000000	0.584906	0.604651	0.577778	1.0	1.000000	0.777778	0.153846	0.090909	0.000000
780	0.045230	0.061652	0.085382	0.062775	0.034350	0.020973	0.033085	0.583333	0.528302	0.511628	0.503704	0.5	0.666667	0.500000	0.115385	0.090909	0.666667
382	0.134857	0.000000	0.253133	0.291998	0.000000	0.000000	0.079536	0.694444	0.000000	0.542636	0.555556	0.0	0.000000	0.777778	0.038462	0.000000	0.333333
1170	0.201214	0.237478	0.261869	0.236503	0.127605	0.085158	0.190828	0.722222	0.566038	0.550388	0.511111	0.5	0.583333	0.722222	0.192308	0.181818	1.000000

df_train.columns

Index(['ridership Barrie', 'ridership Kitchener', 'ridership Lakeshore East',
       'ridership Lakeshore West', 'ridership Milton',
       'ridership Richmond Hill', 'ridership Stouffville',
       'total_train_trips Barrie', 'total_train_trips Kitchener',
       'total_train_trips Lakeshore East', 'total_train_trips Lakeshore West',
       'total_train_trips Milton', 'total_train_trips Richmond Hill',
       'total_train_trips Stouffville', 'Week', 'Month', 'Year'],
      dtype='object')

for st in stations:
    print("---------------------------------------------------------")
    print(st)

    X_train = df_train[f'ridership {st}']
    y_train = df_train.drop(columns = [f'ridership {st}'])
    X_train = X_train.values.reshape(-1,1)
    
    X_test = df_test[f'ridership {st}']
    y_test = df_test.drop(columns = [f'ridership {st}'])
    X_test = X_test.values.reshape(-1,1)
    
    model = LinearRegression()
    model.fit(X_train, y_train)
    r_sq = model.score(X_train, y_train)
    print(f"coefficient of determination (R2) for ridership in {st}: {r_sq} \n")
    
    pred = model.predict(X_test)
#     print(sklearn.metrics.SCORERS.keys())
    print(f"ridership in {st} \n intercept: {model.intercept_} \n") #,  print(f"slope: {model.coef_}")

---------------------------------------------------------
Barrie
coefficient of determination (R2) for ridership in Barrie: 0.5456041019239153 

ridership in Barrie 
 intercept: [-0.00571456  0.05546336  0.05641107 -0.04451814 -0.06144351 -0.01752346
  0.36063978  0.20357201  0.26068896  0.29067713  0.12225861  0.15187642
  0.3347453   0.49601628  0.50685355  0.64636185] 

---------------------------------------------------------
Kitchener
coefficient of determination (R2) for ridership in Kitchener: 0.5499337177626018 

ridership in Kitchener 
 intercept: [ 0.01385063  0.06812625  0.07048068 -0.03208256 -0.04619181 -0.00632009
  0.36653798  0.19192189  0.26474366  0.29446895  0.118598    0.14629704
  0.32975037  0.49574668  0.50629264  0.62024474] 

---------------------------------------------------------
Lakeshore East
coefficient of determination (R2) for ridership in Lakeshore East: 0.5299003721402513 

ridership in Lakeshore East 
 intercept: [-0.05650392 -0.05533666  0.00278735 -0.1013798  -0.12312272 -0.07093549
  0.33814097  0.18872593  0.23463757  0.26904168  0.09759951  0.130835
  0.32250823  0.49215674  0.50057703  0.67120035] 

---------------------------------------------------------
Lakeshore West
coefficient of determination (R2) for ridership in Lakeshore West: 0.4764708528587759 

ridership in Lakeshore West 
 intercept: [-0.04537901 -0.04091711  0.00493538 -0.08490122 -0.10533518 -0.05872361
  0.34327974  0.20597409  0.24139782  0.2719299   0.113469    0.14831997
  0.33287304  0.48479945  0.49264487  0.65858883] 

---------------------------------------------------------
Milton
coefficient of determination (R2) for ridership in Milton: 0.538718449905788 

ridership in Milton 
 intercept: [ 0.04614351  0.03440803  0.09573218  0.09437784 -0.01323931  0.02468805
  0.38515319  0.21633558  0.27881367  0.30670978  0.14128046  0.16803043
  0.34940705  0.50423146  0.5162594   0.62873411] 

---------------------------------------------------------
Richmond Hill
coefficient of determination (R2) for ridership in Richmond Hill: 0.5292378282682253 

ridership in Richmond Hill 
 intercept: [0.0582649  0.046546   0.10588062 0.10318252 0.01296975 0.03632706
 0.39116376 0.2235361  0.28369734 0.31086054 0.14917943 0.17540263
 0.35506801 0.50547923 0.51774157 0.62728369] 

---------------------------------------------------------
Stouffville
coefficient of determination (R2) for ridership in Stouffville: 0.5469105375381922 

ridership in Stouffville 
 intercept: [ 0.02079081  0.01127503  0.07315633  0.07308379 -0.02465224 -0.03985841
  0.37197414  0.20720187  0.2683451   0.29762197  0.12953988  0.1577856
  0.33558373  0.49705506  0.50777433  0.62878011] 

k-fold CV

for st in stations:
    print("---------------------------------------------------------")
    print(st)

    X_train = df_train[f'ridership {st}']
    y_train = df_train.drop(columns = [f'ridership {st}'])
    X_train = X_train.values.reshape(-1,1)
    
    X_test = df_test[f'ridership {st}']
    y_test = df_test.drop(columns = [f'ridership {st}'])
    X_test = X_test.values.reshape(-1,1)
    
    model = LinearRegression()

    r2 = cross_val_score(model, X_train,y_train,scoring='r2',cv=10)
    mean_r2 = np.mean(r2)
    print(f"coefficient of determination (R2) for ridership in {st}: {mean_r2} \n")

---------------------------------------------------------
Barrie
coefficient of determination (R2) for ridership in Barrie: 0.5407156018164897 

---------------------------------------------------------
Kitchener
coefficient of determination (R2) for ridership in Kitchener: 0.5446405643978762 

---------------------------------------------------------
Lakeshore East
coefficient of determination (R2) for ridership in Lakeshore East: 0.5243940620698606 

---------------------------------------------------------
Lakeshore West
coefficient of determination (R2) for ridership in Lakeshore West: 0.46989137157669536 

---------------------------------------------------------
Milton
coefficient of determination (R2) for ridership in Milton: 0.533566630998902 

---------------------------------------------------------
Richmond Hill
coefficient of determination (R2) for ridership in Richmond Hill: 0.5239958791982269 

---------------------------------------------------------
Stouffville
coefficient of determination (R2) for ridership in Stouffville: 0.5418033441563218 

Ridge and Lasso Regression

for st in stations:
    print("---------------------------------------------------------")
    print(st)

    X_train = df_train[f'ridership {st}']
    y_train = df_train.drop(columns = [f'ridership {st}'])
    X_train = X_train.values.reshape(-1,1)
    
    X_test = df_test[f'ridership {st}']
    y_test = df_test.drop(columns = [f'ridership {st}'])
    X_test = X_test.values.reshape(-1,1)
    
    ridge = Ridge()
    parameters = {'alpha':[1e-15,1e-10,1e-8,1e-3,1e-2,1,5,10,20,30,40,50,100]}
    ridge_reg = GridSearchCV(ridge, parameters, scoring='r2', cv=10)
    ridge_reg.fit(X_train,y_train)

    print(f"Best parameters for ridership in {st}: {ridge_reg.best_params_}")
    print(f"Best score for ridership in {st}: {ridge_reg.best_score_}")
    
    pred_ridge=ridge_reg.predict(X_test)

---------------------------------------------------------
Barrie
Best parameters for ridership in Barrie: {'alpha': 0.01}
Best score for ridership in Barrie: 0.540716007842814
---------------------------------------------------------
Kitchener
Best parameters for ridership in Kitchener: {'alpha': 0.01}
Best score for ridership in Kitchener: 0.5446411252243932
---------------------------------------------------------
Lakeshore East
Best parameters for ridership in Lakeshore East: {'alpha': 0.01}
Best score for ridership in Lakeshore East: 0.5243948064907303
---------------------------------------------------------
Lakeshore West
Best parameters for ridership in Lakeshore West: {'alpha': 0.01}
Best score for ridership in Lakeshore West: 0.46989244440754
---------------------------------------------------------
Milton
Best parameters for ridership in Milton: {'alpha': 0.01}
Best score for ridership in Milton: 0.5335669780217649
---------------------------------------------------------
Richmond Hill
Best parameters for ridership in Richmond Hill: {'alpha': 0.01}
Best score for ridership in Richmond Hill: 0.5239961688037555
---------------------------------------------------------
Stouffville
Best parameters for ridership in Stouffville: {'alpha': 0.01}
Best score for ridership in Stouffville: 0.5418038013072269

for st in stations:
    print("---------------------------------------------------------")
    print(st)

    X_train = df_train[f'ridership {st}']
    y_train = df_train.drop(columns = [f'ridership {st}'])
    X_train = X_train.values.reshape(-1,1)
    
    X_test = df_test[f'ridership {st}']
    y_test = df_test.drop(columns = [f'ridership {st}'])
    X_test = X_test.values.reshape(-1,1)
    
    lasso=Lasso()
    parameters={'alpha':[1e-15,1e-10,1e-8,1e-3,1e-2,1,5,10,20,30,40,50,100]}
    lasso_reg=GridSearchCV(lasso,parameters,scoring='r2',cv=5)
    lasso_reg.fit(X_train,y_train)

    print(f"Best parameters for ridership in {st}: {lasso_reg.best_params_}")
    print(f"Best score for ridership in {st}: {lasso_reg.best_score_}")
    
    pred_lasso=lasso_reg.predict(X_test)

---------------------------------------------------------
Barrie
Best parameters for ridership in Barrie: {'alpha': 1e-08}
Best score for ridership in Barrie: 0.5420976438477974
---------------------------------------------------------
Kitchener
Best parameters for ridership in Kitchener: {'alpha': 1e-08}
Best score for ridership in Kitchener: 0.5465688818069757
---------------------------------------------------------
Lakeshore East
Best parameters for ridership in Lakeshore East: {'alpha': 1e-08}
Best score for ridership in Lakeshore East: 0.5261094743605513
---------------------------------------------------------
Lakeshore West
Best parameters for ridership in Lakeshore West: {'alpha': 1e-08}
Best score for ridership in Lakeshore West: 0.4721556113154387
---------------------------------------------------------
Milton
Best parameters for ridership in Milton: {'alpha': 1e-08}
Best score for ridership in Milton: 0.5352207769459096
---------------------------------------------------------
Richmond Hill
Best parameters for ridership in Richmond Hill: {'alpha': 1e-08}
Best score for ridership in Richmond Hill: 0.5257061441010771
---------------------------------------------------------
Stouffville
Best parameters for ridership in Stouffville: {'alpha': 1e-08}
Best score for ridership in Stouffville: 0.5434220868089866